Forecasting

Question 1

What is forecasting?

Answer 1

Forecasting is estimating the future demand for products or services.

Question 2

Why is forecasting important?

Answer 2

Accurate forecasting helps balance supply and demand, avoiding overproduction or underproduction.

Question 3

What are the consequences of underestimated forecasting?

Answer 3

Underestimated forecasting can lead to unsatisfied customers due to insufficient products or services.

Question 4

What are the consequences of overestimated forecasting?

Answer 4

Overestimated forecasting results in excess inventory or staff, leading to increased costs.

Question 5

What are qualitative forecasting methods?

Answer 5

Methods that do not rely on historical data, such as the Delphi method, using expert opinions.

Question 6

What are time series forecasting methods?

Answer 6

Techniques using historical data to predict future trends, e.g., Moving Average, Exponential Smoothing.

Question 7

What is a causal forecasting method?

Answer 7

It analyzes the relationship between different variables to make predictions.
–> Causality on two matrixes

Question 8

What is the role of error in forecasting?

Answer 8

Error measures the difference between actual and forecasted values to assess prediction accuracy.

Question 9

What is mean absolute percentage error (MAPE)?

Answer 9

MAPE measures forecast accuracy as a percentage of error relative to actual sales.

Question 10

What are the components of a time series in forecasting?

Answer 10

-Trend (a product that shows continuous growth )
-seasonality (different levels of demand throughout the year )
-random variations (: we will not see a trend or seasonality but we will have some variation in the data)

Question 11

What is a moving average in forecasting?

Answer 11

The moving average technique is a very simple technique to predict and forecast
–> Applied when there is no trend and no seasonality

Question 12

What is the difference between simple and weighted moving averages?

Answer 12

Weighted averages assign more importance to recent periods, unlike simple averages.

Question 13

What is exponential smoothing?

Answer 13

A forecasting method emphasizing recent data points with weights decreasing exponentially for older data.

Question 14

What is a seasonality index?

Answer 14

A factor quantifying seasonal variation in demand relative to an average value.

Question 15

What are forecasting technics ?

Answer 15

1. Qualitative Methods
2. Time Series Methods
3. Causal Methods

Question 16

How you approach Forecasting different steps (6 steps)

Answer 16

1. Determine the use of the forecast –> What are we trying to predict?
   –> define what are we looking for
2. Identify the items you want to forecast –> outcome variable you want to predict
3. Determine the time horizon –> is it a short time forecast or is it a long time forecast? Are data available to do this in reliable way?
4. Select and build –> determine which kind of model is the best one to predict (different kind of techniques, different kinds of models, you have to choose the good on)
5. Gather data: you have to get data otherwise you can’t predict
6. Validate, good prediction or is it risky

Question 17

What happens if the number of periods used to calculate a Moving Average is decreased?

Answer 17

With a rising trend, underestimation decreases due to a reduced “lag” in the moving average with fewer periods, leading to faster responsiveness to data changes and improved forecasting, though this may not hold for negative trends

Question 18

Weighted Moving Average

Answer 18

takes into account the trend and seasonality
–> Use Weight to weight the most important month

Question 19

CMA (Centered Moving Average)

Answer 19

is a technique used in time series analysis to smooth data and identify trends

Question 20

Season index

Answer 20

is a factor that quantifies the degree of seasonal variation in a time series, helping to identify and adjust for regular, repeating patterns over a specific period (e.g., months, quarters)

Question 21

Season index > 1

Answer 21

Indicates that the period has higher-than-average values (e.g., a month or quarter with higher sales or activity)

Question 22

Season index < 1

Answer 22

Indicates that the period has lower-than-average values (e.g., a slow month or quarter)

Question 23

Season index = 1

Answer 23

The period is at the overall average level

Question 24

Unseasoned value

Answer 24

The actual sales figure for December, say $50,000, which includes the natural seasonal peak due to holiday shopping
–> refers to a raw data point or measurement that has not been adjusted for seasonal variations or recurring patterns that typically occur over regular intervals, such as weekly, monthly, or annually

Question 25

Seasonally adjusted value

Answer 25

After adjusting for the usual holiday sales surge, the adjusted sales might be $35,000, reflecting what the sales would likely be without the seasonal effect

Question 26

Trend-based forecast:

Answer 26

is a method of predicting future values in a time series based on the observed trend in historical data

Question 27

seasoned Trend

Answer 27

is a forecasting model that combines both a trend and seasonal effects to predict future values